University of Rhode Island DigitalCommons@URI

Biological Sciences Faculty Publications Biological Sciences

2020

Seasonal Movements and Habitat Use of Juvenile Smooth Hammerhead Sharks in the Western North Atlantic Ocean and Significance for Management

Ryan K. Logan

Jeremy J. Vaudo

Lara L. Sousa

Mark Sampson

Bradley M. Wetherbee

See next page for additional authors

Follow this and additional works at: https://digitalcommons.uri.edu/bio_facpubs Authors Ryan K. Logan, Jeremy J. Vaudo, Lara L. Sousa, Mark Sampson, Bradley M. Wetherbee, and Mahmood S. Shivji fmars-07-566364 August 30, 2020 Time: 10:2 # 1

ORIGINAL RESEARCH published: 01 September 2020 doi: 10.3389/fmars.2020.566364

Seasonal Movements and Habitat Use of Juvenile Smooth Hammerhead Sharks in the Western North Atlantic Ocean and Significance for Management

Ryan K. Logan1,2*, Jeremy J. Vaudo1,2, Lara L. Sousa3, Mark Sampson4, Bradley M. Wetherbee1,5 and Mahmood S. Shivji1,2*

1 Guy Harvey Research Institute, Nova Southeastern University, Dania Beach, FL, United States, 2 Save Our Seas Foundation Shark Research Center, Nova Southeastern University, Dania Beach, FL, United States, 3 Wildlife Conservation Research Unit, Department of Zoology, The Recanati-Kaplan Centre, University of Oxford, Tubney, United Kingdom, 4 Fish Finder Adventures, Ocean City, MD, United States, 5 Department of Biological Sciences, University of Rhode Island, Kingston, RI, United States Edited by: Yannis Peter Papastamatiou, Florida International University, Upper trophic level predators dramatically impacted by fisheries include the large-bodied United States hammerhead sharks, which have become species of conservation concern worldwide. Reviewed by: James Ketchum, Implementing spatial management for conservation of hammerhead populations Independent Researcher, La Paz, requires knowledge of temporal distribution patterns and habitat use, identification of Mexico essential habitat for protection, and quantification of interactions with human activities. Camrin Braun, University of Washington, There is little such information for the smooth hammerhead shark, Sphyrna zygaena. United States We used fin-mounted satellite tags to examine the movements and habitat use of *Correspondence: juvenile smooth hammerheads, a demographic segment particularly threatened by Ryan K. Logan [email protected]; exploitation. Six sharks were tagged off the US mid-Atlantic and tracked for 49–441 [email protected] days (mean 187 ± 136 days). Sharks consistently showed area-restricted movements Mahmood S. Shivji within a summer core area in waters of the New York Bight and a winter core area off [email protected] Cape Hatteras, North Carolina, with directed movements between them in autumn. Specialty section: There was high overlap of shark winter core area use and the Mid-Atlantic Shark This article was submitted to Area (MASA) – a 7 month per year, bottom-longline fishery closure – indicating that Marine Megafauna, a section of the journal this area closure offers seasonal reduction in fishing pressure for this species. Based Frontiers in Marine Science on timing of shark movements and the MASA closure, protection for juvenile smooth Received: 27 May 2020 hammerheads may be increased by beginning the closure period 1 month earlier than Accepted: 12 August 2020 Published: 01 September 2020 currently scheduled. Generalized additive mixed models revealed that area-restricted Citation: movements of sharks in their summer and winter core areas coincided with high primary Logan RK, Vaudo JJ, Sousa LL, productivity, and elevated sea surface temperature. Consistency in use of summer and Sampson M, Wetherbee BM and winter core areas suggests that the coastal waters of the New York Bight and Cape Shivji MS (2020) Seasonal Movements and Habitat Use Hatteras, North Carolina could be considered for Essential Fish Habitat designation for of Juvenile Smooth Hammerhead this species. This study reveals the first high resolution movements and habitat use for Sharks in the Western North Atlantic Ocean and Significance smooth hammerheads in the western North Atlantic to inform management planning for for Management. this population. Front. Mar. Sci. 7:566364. doi: 10.3389/fmars.2020.566364 Keywords: Sphyrna zygaena, movement ecology, behavior, conservation, satellite telemetry

Frontiers in Marine Science| www.frontiersin.org 1 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 2

Logan et al. Smooth Hammerhead Shark Seasonal Movement

INTRODUCTION Smooth hammerheads have a circumglobal distribution in coastal and oceanic waters and occupy a wider latitudinal range The rapid expansion of elasmobranch fisheries and trade globally than other sphyrnids (Compagno, 1984). Catch records from are principal drivers of population decline for many shark species a variety of locations suggest that juveniles and sub-adults (Dulvy et al., 2014). Some species, such as the large-bodied [<265 cm total length (TL)] are more common in inshore waters hammerhead sharks (great – Sphyrna mokarran, scalloped – over coastal shelves, with larger individuals (>265 cm TL) found S. lewini, and smooth – S. zygaena hammerheads), are especially more frequently offshore (Diemer et al., 2011; Clarke et al., 2015; vulnerable to fishing pressure because of their slow rates of Francis, 2016; Deacy et al., 2020). This species is capable of long population growth (Cortés et al., 2010) and high at-vessel and distance movements (e.g., 6,610 km over 150 days; Santos and post-release mortality due to elevated stress response to capture Coelho, 2018), but also shows high levels of resident behavior (Morgan and Carlson, 2010; Eddy et al., 2016; Gallagher and within restricted areas (at least 80 days; Diemer et al., 2011). Klimley, 2018). In addition, hammerhead sharks are taken in To date, just two studies have reported on the finer-scale large numbers because of the high demand for their superior- movements of smooth hammerheads via satellite telemetry. quality fins (large size and high ceratotrichia count) in the global Horizontal and vertical movements of juveniles in the temperate shark fin trade (Abercrombie et al., 2005; Clarke et al., 2006a,b; western South Pacific appear to vary seasonally (Francis, Cardeñosa et al., 2018). While population declines of the large- 2016); in contrast, juvenile and adult smooth hammerheads bodied hammerhead shark species complex is thought to be in the tropical eastern Atlantic demonstrated relatively stable largely driven by declines of scalloped hammerheads (Jiao et al., temporal diving behavior (Santos and Coelho, 2018). These 2011), low catch rates in various parts of the world for all species studies illustrate the possibility that movements and behavior suggest a significant historical decline in the abundance of all of smooth hammerheads may be influenced by thermal large-bodied hammerhead sharks (Baum et al., 2003; Baum and heterogeneity of their environment. Very little information on Blanchard, 2010; Ferretti et al., 2010). relationships between other environmental variables (primary Fishery exploitation of smooth hammerhead sharks via productivity, oceanic fronts, water depth) and movements of targeting or bycatch has been identified as the major threat smooth hammerheads exists (Couto et al., 2018), although to this species, particularly for juveniles (Casper et al., 2009; these variables have been shown to greatly influence movement Cortés et al., 2010; Miller, 2016). The conservation of this and habitat selection of highly mobile sharks (Block et al., species is an international priority, e.g., Vulnerable listing on 2011; Queiroz et al., 2016; Vaudo et al., 2017). Understanding the International Union for Conservation of Nature (IUCN) Red interactions between patterns of habitat use and environmental List (Casper et al., 2009); Appendix II listing on the Convention variables would contribute toward construction of habitat on International Trade in Endangered Species of Wild Fauna models and an improved ability to predict the distribution and Flora (CITES); Appendix II listing on the Convention of smooth hammerheads under climate change scenarios, as on Migratory Species of Wild Animals (CMS). Furthermore, well as reveal potential interactions with human activities an ecological risk assessment of sharks caught in Atlantic throughout their range. pelagic longline fisheries highlighted smooth hammerheads as Given minimal information on the spatial ecology of smooth a species in urgent need of biological data necessary for stock hammerheads in general and conservation concerns for this assessment (Cortés et al., 2010). However, conservation-relevant species, our goal was to quantify habitat use and horizontal data on many aspects of the biology of smooth hammerheads movements of juvenile smooth hammerheads in the western are extremely limited, including information relating to their North Atlantic Ocean via satellite tag telemetry. Only very movements, seasonal distributions and habitat use (Miller, 2016; coarse scale information exists on movements of smooth Gallagher and Klimley, 2018). hammerheads in this region, obtained from the recapture of just Achieving sustainable populations of fishery exploited species seven individuals out of 269 (0.02%) tagged with conventional is critically dependent on the recruitment of immature identification tags over 52 years (Kohler and Turner, 2019). Our individuals. Thus, identification and conservation of essential study objectives were to: (1) determine seasonal movements and habitat for juvenile and sub-adult sharks is of paramount distribution patterns; (2) identify core areas of habitat use; (3) importance, requiring an understanding of this key demographic evaluate the potential of the MASA seasonal closure for providing segment’s spatiotemporal patterns of occurrence and associated protection from fishing pressure, and (4) investigate relationships oceanic environmental drivers (Kinney and Simpfendorfer, 2009; between movement behavior and environmental conditions, for Schlaff et al., 2014). Equipped with adequate information on how juvenile smooth hammerheads. environmental parameters influence seasonal movements, spatial management measures such as temporal closures of targeted areas can be enacted to promote recovery of overfished stocks. MATERIALS AND METHODS For example, based on understanding of temporal and spatial habitat use of dusky sharks (Carcharhinus obscurus), the Mid- Capture and Tagging Atlantic Shark Area (MASA) – a region closed to bottom longline Between 22 July 2016 and 9 September 2017 six female, juvenile fishing for 7 months each year – was established in 2005 as a smooth hammerhead sharks were caught via rod and reel off means of reducing fishing mortality and enhancing recovery of the coast of Ocean City, Maryland United States (38.1◦ N, 74.5◦ this species (NMFS, 2009). W). Sharks were brought on board the fishing vessel where a

Frontiers in Marine Science| www.frontiersin.org 2 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 3

Logan et al. Smooth Hammerhead Shark Seasonal Movement

saltwater hose was inserted into the mouth to irrigate the gills is high (resident) or low (transient) the classification can be and then the sharks were measured, sexed, and fitted with a confidently assessed. Consequently, following Breed et al.(2009), satellite-linked radio tag (SPOT-196 tag; Wildlife Computers, we classified proportions ≤0.3 as transient, ≥0.7 as resident, and Redmond, WA, United States) on the dorsal fin. These tags proportions of 0.3–0.7 as uncertain. The hSSM was fit using the directly communicate with the Argos tracking system1 when bsam package (Jonsen et al., 2015) in R. the shark’s dorsal fin breaks the sea surface exposing the tag Using the hSSM positions, a seasonal utilization distribution to air, providing an estimated position (latitude and longitude) (UD) was calculated for all sharks pooled across the and an associated location class. Location class is determined by meteorological seasons (summer: June–August, autumn: the number of transmissions received and the number of Argos September–November, winter: December–February, and spring: satellites receiving transmissions, and categorized from most to March–May) using the adehabitat package in R (Calenge, least accurate as 3, 2, 1, 0, A and B. Estimated errors (1 SD) for 2006). The UD estimate was calculated following methods each location class are LC 3: < 250 m, LC 2: 250–500 m; LC 1: described in Vaudo et al.(2017). 500–1500 m, and LC 0: > 1500 m; there is no spatial estimate of To investigate vertical diel behavior in the absence of accuracy for LC A and B (CLS, 2016). The two tags deployed in transmitted depth data, we used successful Argos transmissions 2016 were programmed to transmit for 1 h every other hour; the as a proxy for surfacing behavior since locations are only four tags deployed in 2017 were programmed to transmit for 1 h obtained when sharks are at the surface (Doyle et al., 2015). every 4 h to try to obtain longer duration tracks. Using the Argos Satellite Pass Prediction tool2, satellite pass data was obtained for all six available satellites from June 2017– Movements, Distribution, and Behavior September 2018. Because each satellite can simultaneously detect Because Argos positions of sharks varied in temporal frequency all transmitters within an approximately 5,000 km diameter and spatial accuracy, we obtained standardized positions circle below it (CLS, 2016), satellite pass data was obtained ◦ ◦ (hereafter “positions”) at 12 h intervals that were comparable for 40 N and 74 W, which encompassed all shark positions between individuals and over time by processing Argos locations received. Because the number of satellites passing overhead using a behavioral switching state-space model (SSM) within a varies by hour of the day (in effect increasing the amount of Bayesian framework developed by Jonsen et al.(2005). Since listening effort when more satellites are present; Supplementary parameter estimation is improved when conducted jointly across Figure S2), surfacing behavior was determined by summing the multiple individual datasets (Jonsen, 2016), we produced most number of Argos locations obtained per shark per hour (Eastern probable tracks using a hierarchical joint estimation model Standard Time), and dividing by the cumulative amount of time (hSSM) that produced temporally regular positional estimates that all satellites were overhead during each hour (in general, based on the Argos location class, mean turning angle, and each satellite takes roughly 10 min to pass over a stationary autocorrelation in speed and direction. Previous research has object). The resulting value represents a standardized number of shown that the accuracy of the hSSM parameter estimates Argos locations per hour of satellite coverage (hereafter termed declines in response to outlier locations (from poor quality “surfacing index”), providing information on temporal patterns satellite positions) and long gaps in detection data (Bailey et al., of surfacing, regardless of the number of satellites overhead. The 2008); therefore, prior to fitting hSSMs, each track was filtered surfacing index (square root transformed) was compared among using the argosfilter package (Freitas et al., 2008) in R Core hours of the day using a linear mixed effects (LME) model as Team(2014) with parameters listed in Vaudo et al.(2017). surfacing index ∼ hour + ID, where surfacing index was the To reduce spurious results associated with long detection gaps, response variable, hour of day was the explanatory variable and tracks were broken into multiple segments when gaps between shark ID was a random factor using the lmer function in the Argos locations were >10 days. Resulting segments <20 days in lme4 R package (Bates et al., 2014). Tests of multiple comparisons duration were excluded from the hSSM (Block et al., 2011). Given were obtained using the glht function in the multcomp package that 84.7% of gaps between positions in our tracks were <12 h (Hothorn et al., 2008). This analysis of diel surfacing behavior (Supplementary Figure S1), we used a time step of 12 h in the was limited to Sharks 3–6 because satellite pass data is only hSSM to produce two positions per day for each shark. retained by the Argos system for 1 year and this analysis was The hSSM model was fit by running two Markov Chain not undertaken until 2018; thus, satellite pass data could only be Monte Carlo (MCMC) chains in parallel for a total of 60,000 obtained for the four sharks tagged in 2017. samples, with the first 50,000 being discarded as burn-in, and the remaining 10,000 samples thinned by retaining every 10th Environmental Variables sample to reduce autocorrelation (n = 1,000 per chain). Each Water depth (m) and sea surface temperature (SST; ◦C) MCMC iteration provides not only a most probable track but also values were obtained using the NOAA ETOPO1 Global assigns each estimated location to one of two possible behavior Relief Model (one arc-minute resolution) and the Multi- modes (resident and transient). The final estimated track is scale Ultra-high Resolution (MUR) SST dataset3 (0.01◦ the average of all 2,000 MCMC samples and the final output resolution), respectively, using the “xtractomatic” package for each behavioral state represents the proportion of samples in R (Mendelssohn, 2017). SST gradient (a proxy for temperature for a given position classified as resident (MCMC diagnostics fronts) was calculated as the maximum difference in SST given in Supplementary Figure S5). When the proportion 2https://argos-system.clsamerica.com 1www.argos-system.org 3http://mur.jpl.nasa.gov/

Frontiers in Marine Science| www.frontiersin.org 3 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 4

Logan et al. Smooth Hammerhead Shark Seasonal Movement

across a moving window of a 15 × 15 grid cell matrix (median = 2.5 h). The number of days with Argos locations for (totaling ∼0.15◦ area covered) using the raster package each shark ranged from 46 to 263 days (mean 131 ± 72.5 days), (Hijmans et al., 2017) in R. Finally, using the rerddap resulting in a total of 786 days with locations out of 1121 days at (Chamberlain et al., 2019) and rerddapXtracto4 packages in liberty (mean 187 ± 136 days) (Table 1). Among all sharks, this R, we obtained 8-day composite primary productivity (PP) equates to being detected on 77.2 ± 0.1% of days at liberty. Once (mg C/m2/day; 0.0125◦ resolution) data from the National the Argos locations were filtered and standardized to a 12 h time Aeronautics and Space Administration’s (NASA) Aqua satellite interval using the hSSM, positions were removed for days lacking with its Moderate Resolution Imaging Spectroradiometer an Argos location. As a result, 1,531 positions remained, which sensor (MODIS-Aqua). served as the basis of subsequent analyses. Generalized additive mixed models (GAMMs) were used Most individuals displayed similar movements and habitat use to determine the best environmental predictors of smooth throughout the course of the study (Figure 1A). In general, the hammerhead shark resident behavior. Prior to inclusion in hSSM indicated that sharks were resident in shallow water off the global model, univariate models were constructed with southern Long Island, New York during the summer, with some potential environmental predictors standardized by their mean southern movement to the waters off New Jersey, Delaware and and standard deviation, and tested using a likelihood ratio test. Maryland in late summer. During autumn, directed southern Significant predictor variables were then tested for collinearity movements through the mid-Atlantic region were common to using a Pearson’s rank correlation matrix (Zuur et al., 2009) all sharks, showing little affinity to any one region in the area, as and all non-spatial combinations were <0.7 (Supplementary indicated by the observation that 57% of positions were classified Figure S3). The proportion of behavioral states categorized as as transient during autumn. During winter and early spring, resident for each position by the hSSM was used as the response sharks displayed area restricted movements, primarily focused variable and all predictor variables were included in the global near the southeastern outer banks of Pamlico Sound, North model. The model was run using a Gaussian response distribution Carolina (Figure 1A and Supplementary Figure S4). and identity link. The importance of various combinations One shark (shark #5), tagged 17 June 2017 was tracked for of autocorrelation structures was tested while holding other 441 days (Figure 1b and Table 1). This shark was tagged near variables constant. Similarly, to determine the best random Ocean City, Maryland and remained there for almost 2 weeks effects structure aimed at accounting for any temporal effect after tagging, then moved north into the New York Bight in (e.g., increasing temporal gaps between positions since tagging early July, where it remained until September when it began a 3 or season) or individual effect imposed on the sharks’ behavior, month journey south reaching the area off Cape Hatteras, North we considered shark ID, season and days at liberty as possible Carolina on 15 November. It remained in this area until 3 May random effects. The performance of the final model output was 2018 and was not detected again until 9 July 2018 off New Jersey; assessed using the C index, where values closer to 1 indicate the shark then moved to waters of the New York Bight until better performance, and the corresponding Somers’ Dxy rank 1 September 2018. Another individual (Shark #1) moved south correlation, which is a measure of ordinal association between the from the New York Bight similarly to other sharks, but continued response and predictor variables (Lea et al., 2018). moving south east Cape Hatteras eventually reached the coast of central Florida on 12 December 2016 at the time of the last detection 144 days after tagging (Figure 1a). RESULTS Seasonal utilization distributions showed similar seasonal movements as indicated by the hSSM (Figure 2). Core areas (50% The six juvenile smooth hammerheads TL (mean ± SD) UD) of the seasonal distributions were primarily centered in the 184.2 ± 18.5 cm were tracked for periods of 49–441 days and New York Bight in the summer, expanded southward during the generated a total of 3,488 Argos locations. The number of Argos autumn as sharks moved south, and were concentrated off Cape locations d−1 ranged from 0 to 21 (mean 3.1 ± 3.3). The Hatteras, North Carolina in the winter (Figure 2). The individual mean time interval between Argos positions was 7.7 ± 33.8 h tracked for greater than 1 year moved north in late April/early May, in a similar manner to the northward movements of sharks 4https://github.com/rmendels/rerddapXtracto shortly after tagging off Ocean City, Maryland (Figures 1b, 2d).

TABLE 1 | Summary information for SPOT tag deployments on juvenile smooth hammerhead sharks.

Shark ID TL (cm) Sex Date tagged Tagging location Days detected Track duration (days) Track distance (km) Argos locations day−1

1 221 F 22-Jul-16 38.22, −75.03 126 144 3305.2 6.6 ± 4.9 2 183 F 18-Sep-16 38.27, −74.8 118 155 2554.1 3.7 ± 3.6 3 163 F 4-Jun-17 37.96, −74.63 139 217 4359.4 2.4 ± 2.3 4 173 F 12-Jun-17 37.98, −74.75 94 115 2252.2 2.6 ± 1.9 5 190 F 17-Jun-17 37.95, −74.71 263 441 7319.5 2.2 ± 2.5 6 175 F 13-Sep-17 38.25, −74.8 46 49 1345.8 3.0 ± 2.2

Track distance reflects the sum of distances between estimated track positions. TL: shark total length.

Frontiers in Marine Science| www.frontiersin.org 4 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 5

Logan et al. Smooth Hammerhead Shark Seasonal Movement

FIGURE 1 | (a) Twelve-hour position estimates for six juvenile smooth hammerhead sharks determined by a hierarchical Bayesian state space movement model (hSSM). (b) Shark #5 tracked 441 days revealing a complete migration. Black star, tagging location; NY, New York; NJ, New Jersey; MD, Maryland; VA, Virginia; NC, North Carolina; SC, South Carolina; GA, Georgia; FL, Florida.

For sharks with transmissions extending to the winter and greatest number of satellite passes in the study area (all satellites spring of the year following tag deployment (n = 4), 96.7% (315 combined = 794 passes, Supplementary Figure S2), resulting in of 326) of locations fell within the boundaries of the Mid-Atlantic 165.3 h of listening time, which yielded an overall surfacing index Shark Area (MASA) off North Carolina (Figure 3); 101 (32%) of of 4.23 for all sharks combined. However, although the 0400– these positions occurred during the month of December, when 0500 h block only had 40.8 total h of satellite listening time, it the area is open to commercial bottom longline fisheries (closure had the highest surfacing index of 4.35 for all sharks combined period: 1 January–31 July). Positional data was not available to (i.e., on average, there was roughly one position per shark per determine when shark #5 (the individual tracked for >1 year) left hour of satellite coverage just before and just after dawn and dusk, the MASA (Figure 1b), and transmissions from all other sharks respectively; Figure 4). stopped prior to exiting the MASA, so time spent within the After testing the importance of various combinations of MASA could not be assessed. autocorrelation structures while holding other variables constant, Because tags (sharks 3–6) deployed in 2017 were programmed we found that the GAMM without an autocorrelation term was to transmit just one out of every 4 h, diel vertical behavior is deemed more robust with better wAIC and 1AIC (wAIC = 0.31; only described for the hours tags were set to transmit (0000– Supplementary Table S1); thus, no autocorrelation structure 0100, 0400–0500, 0800–0900, 1200–1300, 1600–1700, and 2000– was used in the final model. Additionally, wAIC and 1AIC 2100 h). Significant fixed effects for the 0400–0500 and 2000– revealed that treating shark ID and season as random effects 2100 h blocks (0400–0500 LME Estimate = 0.25, SE = 0.07, resulted in the most parsimonious model, thus, days at liberty t = 3.4, p = 0.004; 2000–2100 LME Estimate = 0.23, SE = 0.07, was not included (Supplementary Table S2). The final GAMM t = 3.1, p = 0.007) indicated that surfacing index varied over the predicting residency behavior explained 34% of the sample 24 h diel period, and multiple comparisons revealed that sharks variance (C index = 0.72, Dxy = 0.45, SD = 0.001, n = 1432). surfaced most frequently just before dawn, at midday, and just Mean SST, log of primary productivity and water depth were after dusk (Figure 4). The total number of Argos locations for included in the best fit model, while SST gradient was removed all sharks pooled was greatest during the time interval 0400– given its lack of significance (p = 0.2) and improved model 0500 and 2000–2100 h (308 and 700 total Argos locations, fit after removal (1AIC = 2.4, wAIC = 0.76). Model output respectively). The 2000–2100 h time interval coincided with the indicated that most of the variation in the observed resident

Frontiers in Marine Science| www.frontiersin.org 5 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 6

Logan et al. Smooth Hammerhead Shark Seasonal Movement

FIGURE 2 | Seasonal utilization distributions (UD) for juvenile smooth hammerhead sharks overlaid on mean seasonal sea surface temperature (SST) during the tracking period. Seasons are summer (a; June–August), autumn (b; September–November), winter (c; December–February) and spring (d; March–May). Solid line is the 95% UD, dashed line is the 75% UD and dotted line is the 50% UD. N refers to the number of individuals that were analyzed in each season. Because only one individual represents the spring locations, points of locations are shown. Gray contour lines represent depth contours from 100 to 1000 m depth.

behavior was attributable to geographical location, followed (Figure 5B)], and inshore neritic waters (<100 m; Figure 5C). by primary productivity concentration and depth (Table 2). In addition, SST of ∼18, 23 and >26◦C resulted in increased Probability of displaying resident behavior was highest at probability of sharks displaying resident behavior as these latitudes associated with the New York Bight (>40◦ N), high represented the temperatures experienced in core habitat areas primary productivity concentration [7.82 log(mg C/m2/day) (Figure 5A and Table 2).

Frontiers in Marine Science| www.frontiersin.org 6 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 7

Logan et al. Smooth Hammerhead Shark Seasonal Movement

FIGURE 3 | Locations of juvenile smooth hammerhead sharks (n = 4 for all groupings) within the mid-Atlantic Shark Area (MASA) during the closure period (January 1–July 1; orange points), and outside of the closure period (black points). Nearly all December locations (green points) fall within the MASA boundary but not within the closure period.

DISCUSSION (<10 m) waters (Francis, 2016; Santos and Coelho, 2018), potentially making them good candidates for SPOT tags which We provide the first detailed view of the movement dynamics only transmit data when exposed to air. Indeed, the sharks of smooth hammerhead sharks in the western North Atlantic. tracked here were detected on average 3.1 ± 3.3 times per Previous work has reported that smooth hammerhead sharks day, and 84.7% of Argos locations occurred within 12 h of a spend a large proportion of their time in surface or near-surface previous location. The high frequency of satellite transmissions and Argos locations allowed for reconstruction of smooth hammerhead movements at a much higher resolution than has previously been described. The sharks we tracked in the western North Atlantic Ocean displayed consistent seasonal movements between core areas of activity off Long Island, New York in summer and off Cape Hatteras, North Carolina in winter. Although seasonal movements of this species have been hypothesized previously based on surface sightings (Couto et al., 2018) and fisheries catch per unit effort data (Santos and Coelho, 2019) in the eastern North Atlantic, the telemetry results here provide a direct, fishery independent demonstration of this behavior by smooth hammerheads. Based on environmental characteristics of the core areas, sea surface temperature and productivity appear to be important drivers of their seasonal movement patterns, as has been demonstrated in other highly migratory marine megafauna (Weng et al., 2008; Block et al., 2011; Curtis et al., 2014; Kajiura FIGURE 4 | Surfacing index (number of locations per hour of satellite time) of and Tellman, 2016; Vaudo et al., 2017). juvenile smooth hammerhead sharks. Vertical dashed lines and shading Seasonal movements and habitat use in other hammerhead represent the minimum, mean and maximum times of sunrise and sunset species have been documented, but thus far suggest they experienced by sharks. Bars labeled with different letters differ at α = 0.05, are driven more so by foraging or reproduction, rather and bars without letters were not included in statistical comparisons. Data only includes four sharks tagged in 2017. EST, Eastern Standard Time. than dynamic oceanographic processes. For example, seasonal changes in abundance of scalloped hammerheads at offshore

Frontiers in Marine Science| www.frontiersin.org 7 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 8

Logan et al. Smooth Hammerhead Shark Seasonal Movement

TABLE 2 | GAMM output for juvenile smooth hammerhead resident behavior in productivity, presumably tied to prey availability (Ware and relation to environmental variables. Thomson, 2005; Priede and Miller, 2009). Stomach contents show Variable edf Ref.df F p-value that the major prey of smooth hammerheads is cephalopods (mainly ommastrephid squid) and small schooling fishes (Smale, s(SST) 6.8 6.8 4.2 <0.001 1991; Rogers et al., 2012; Bornatowski et al., 2014). The longfin s[log(Primary Productivity)] 3.9 3.9 25.3 <0.001 squid Doryteuthis pealeii and shortfin squid Illex illecebrosus s(Depth) 1 1 25.4 <0.001 are the most common species of squid in the western North te(Lon, Lat): Summer 10 10 42.1 <0.001 Atlantic from Georges Bank to Cape Hatteras, and both species te(Lon, Lat): Autumn 11.6 11.6 19.4 <0.001 undergo seasonal spawning migrations at northern and inshore te(Lon, Lat): Winter 4.2 4.2 11.9 <0.001 locations in late spring/early summer and deeper, southern te(Lon, Lat): Spring 4.1 4.1 19.5 <0.001 locations along the continental shelf edge in late autumn/early winter (Dawe et al., 2007; Jereb and Roper, 2010). Seasonal movements and aggregations in relation to high prey abundance islands in the eastern tropical Pacific have been suggested has been reported in several species of sharks (Klimley et al., as possibly related to movements for reproductive purposes 1992; Heyman et al., 2001; Mourier et al., 2016), however, and/or parturition, but currents and chlorophyll concentrations little information exists on smooth hammerhead diet in the may also play a role in long term movements (Bessudo study region to determine if they are taking advantage of et al., 2011; Ketchum et al., 2014; Nalesso et al., 2019). this potential resource. Nevertheless, spawning and seasonal Wells et al.(2018) found scalloped hammerhead movements movements of these squid in the western North Atlantic in the northern Gulf of Mexico to be primarily driven by spatially and temporally overlap with core areas used by smooth static bathymetric features rather than dynamic environmental hammerheads tracked in this study. variables and did not observe any seasonal patterns in shark The diel surfacing behavior patterns of smooth hammerheads movements. Furthermore, great hammerhead repeated seasonal tracked in our study may also be related to foraging. Highest presence and residency within the Bahamas is also believed surfacing indices were recorded shortly before dawn and after to be related to reproduction or foraging, rather than climatic dusk, similar to the pattern observed in a juvenile smooth processes (Guttridge et al., 2017). However, sharks tracked here hammerhead (139 cm TL) tracked off the coast of New Zealand represent the juvenile to sub-adult size class of this species, (Francis, 2016). Francis(2016) also reported diel differences and physiological tolerances to environmental conditions vary in depth distribution of another juvenile smooth hammerhead across ontogeny and may result in juveniles selecting different tracked with a popup satellite transmitter, with a shallower habitats than adults (Grubbs, 2010). Given that only one smooth distribution at night compared to daylight hours. In contrast, hammerhead in our study was tracked for over a full year, it scalloped hammerhead sharks have been observed to remain in remains unclear how typical seasonal migratory behavior and shallow waters during the day and dive at night presumably environmental driven movement is in this species throughout its to forage (Klimley and Nelson, 1984; Hoffmayer et al., 2013), geographic and size range. or show continuous deep diving behavior throughout the 24 h Seasonal movement patterns of smooth hammerheads along cycle (Spaet et al., 2017). Similarly, Santos and Coelho(2018) the US East Coast was characterized by resident behavior found that similarly sized smooth hammerheads to those in during the summer and late winter/early spring. The timing this study [T-test; T(6.5) = 1.1, p = 0.3] tracked using depth of resident behavior coincided with increased levels of primary and temperature archival transmitters off the west coast of

FIGURE 5 | Relationship of sea surface temperature (SST; A), log of primary productivity (B) and seafloor depth (C) with resident behavior exhibited by juvenile smooth hammerhead sharks. Values with 95% confidence intervals that do not overlap with 0 (red line) indicate either increased (positive values) or decreased probability of resident behavior (negative values; transient behavior). Black ticks along the x-axis represent the distribution of the independent variable values examined. Note y-axis scales differ.

Frontiers in Marine Science| www.frontiersin.org 8 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 9

Logan et al. Smooth Hammerhead Shark Seasonal Movement

equatorial Africa occupied deeper, cooler water during the night Successful management of populations is dependent on the compared to day. Our findings contrast somewhat with those survival of young individuals and recruitment to reproductive of Santos and Coelho(2018); however, this difference may be stock; therefore, understanding movement patterns, habitat use an artifact of study location, where sharks tracked in Santos and EFH of juveniles is vital. In this study we have identified and Coelho(2018) were experiencing temperatures at depth both winter and summer core areas of concentrated activity for several degrees warmer (26–27◦C) than SSTs observed here in juvenile smooth hammerheads in the western North Atlantic, the temperate western North Atlantic (mean 21.7 ± 2.2◦C). In as well as pathways traveled between those seasonal core addition, the surfacing index presented here is limited to when areas. In addition, environmental conditions associated with a sharks’ dorsal fin breaks the surface and appropriate satellite resident behavior within these core areas and timing of directed coverage is overhead, so patterns observed here may not be fully movements between them enables improved ability to predict representative of smooth hammerhead diel depth distribution in inter- and intra-annual distribution of smooth hammerheads, the western North Atlantic. and how this may change over time with changing environmental Decreasing population trends of smooth hammerheads have conditions (e.g., increasing sea surface temperatures). These prompted conservation listings (e.g., IUCN, CITES, CMS) and advances in understanding patterns of distribution and habitat calls for additional management. Because of high at- vessel use of juvenile smooth hammerheads in the western North (Coelho et al., 2012) and estimated post-release mortality of Atlantic are directly applicable to effective management of smooth hammerheads caught in fisheries (Braccini et al., 2012), this demographic component of their population. Future work reducing exposure to capture rather than relying on release after should include studying the movement ecology of adult smooth capture is a more effective management method for reduced hammerhead sharks of both sexes since their movements fishing mortality. While acknowledging that our inferences and habitat use patterns are likely to be different from are based on the four animals with long enough tracks, the those of juveniles. consistent finding of the winter core area of activity largely falling within the boundaries of the MASA management zone during winter and spring, and high proportion of transmissions DATA AVAILABILITY STATEMENT occurring within the MASA during the shark bottom longline fishery closure period (1 January–31 July), suggests the potential The raw data supporting the conclusions of this article will be of the MASA for reducing fishing mortality of this species. made available by the authors, without undue reservation, to any Furthermore, as reported for sand tiger sharks (Carcharias qualified researcher. taurus)(Teter et al., 2015), the smooth hammerhead spatial and temporal patterns of movement suggest that beginning the MASA closure on 1 December, rather than 1 January, would ETHICS STATEMENT provide additional and extended protection from commercial fisheries for this species also. The animal study was reviewed and approved by the Nova Though there was some individual variability in movements Southeastern University IACUC #DB1. of smooth hammerheads tracked in our study with a limited number of individuals, the high degree of spatial and temporal AUTHOR CONTRIBUTIONS consistency demonstrated by the sharks in use of both summer and winter core areas as well as behaviors associated with BW and MSS designed and implemented the research. MS led the foraging suggest that the coastal waters of the New York Bight fieldwork. RL led the analysis of the data with assistance from LS and Cape Hatteras, North Carolina could be considered for and JV. RL led the writing of the manuscript with assistance from designation of Essential Fish Habitat (EFH) for this species in the all authors. western North Atlantic, an important designation for protection consideration in U.S. fisheries management practices (NMFS, 2009)5. Seasonal movement between southern areas of increased FUNDING presence in winter and northern areas of concentrated activity in summer have been reported for other species of sharks in the This work was supported by grants to MSS from the western North Atlantic, including sandbar sharks Carcharhinus Guy Harvey Ocean Foundation (GHOF18-1), Save Our Seas plumbeus (Grubbs et al., 2007; McCandless et al., 2007; Conrath Foundation (SOSF157), Shark Foundation/Hai-Stiftung (HF-19- and Musick, 2008), dusky sharks Carcharhinus obscurus (Musick 2), the Levitetz Family Foundation (LFF-19), and a scholarship to and Colvocoresses, 1986), sand tiger sharks Carcharias taurus RL from Fish Florida. (Teter et al., 2015) and white sharks Carcharodon carcharias (Curtis et al., 2018), and has led to delineation of nurseries and designation of EFH for several of these species (NMFS, 2009). SUPPLEMENTARY MATERIAL Likely due to the lack of available data, there is currently no EFH in U.S. waters for smooth hammerhead sharks. The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmars.2020. 5https://www.fisheries.noaa.gov/resource/map/essential-fish-habitat-mapper 566364/full#supplementary-material

Frontiers in Marine Science| www.frontiersin.org 9 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 10

Logan et al. Smooth Hammerhead Shark Seasonal Movement

REFERENCES pelagic longline fishery in the Atlantic Ocean. Aquat. Living Resour. 25, 311– 319. doi: 10.1051/alr/2012030 Abercrombie, D. L., Clarke, S. C., and Shivji, M. S. (2005). Global-scale Compagno, L. J. (1984). Sharks of The World: An Annotated and Illustrated genetic identification of hammerhead sharks: application to assessment of the Catalogue of Shark Species Known to Date, pt. 2: Carcharhiniformes. international fin trade and law enforcement. Conserv. Genet. 6, 775–788. doi: Rome: FAO. 10.1007/s10592-005-9036-2 Conrath, C. L., and Musick, J. A. (2008). Investigations into depth and Bailey, H., Shillinger, G., Palacios, D., Bograd, S., Spotila, J., Paladino, F., et al. temperature habitat utilization and overwintering grounds of juvenile (2008). Identifying and comparing phases of movement by leatherback turtles sandbar sharks, Carcharhinus plumbeus: the importance of near shore North using state-space models. J. Exp. Mar. Biol. 356, 128–135. doi: 10.1016/j.jembe. Carolina waters. Environ. Biol. Fishes 82, 123–131. doi: 10.1007/s10641-007- 2007.12.020 9263-0 Bates, D., Maechler, M., Bolker, B., and Walker, S. (2014). lme4: Linear mixed- Cortés, E., Arocha, F., Beerkircher, L., Carvalho, F., Domingo, A., Heupel, M., effects models using Eigen and S4. R package version 1. doi: 10.18637/jss.v0 et al. (2010). Ecological risk assessment of pelagic sharks caught in Atlantic 67.i01 pelagic longline fisheries. Aquat. Living Resour. 23, 25–34. doi: 10.1051/alr/200 Baum, J. K., and Blanchard, W. (2010). Inferring shark population trends from 9044 generalized linear mixed models of pelagic longline catch and effort data. Fish. Couto, A., Queiroz, N., Ketchum, J. T., Sampaio, E., Furtado, M., Cid, A. A., Res. 102, 229–239. doi: 10.1016/j.fishres.2009.11.006 et al. (2018). Smooth hammerhead sharks (Sphyrna zygaena) observed off the Baum, J. K., Myers, R. A., Kehler, D. G., Worm, B., Harley, S. J., and Portuguese southern coast. Environ. Biol. Fishes 101, 1261–1268. doi: 10.1007/ Doherty, P. A. (2003). Collapse and conservation of shark populations s10641-018-0773-8 in the Northwest Atlantic. Science 299, 389–392. doi: 10.1126/science.107 Curtis, T. H., Metzger, G., Fischer, C., McBride, B., McCallister, M., Winn, L. J., 9777 et al. (2018). First insights into the movements of young-of-the-year white Bessudo, S., Soler, G. A., Klimley, A. P., Ketchum, J. T., Hearn, A., and Arauz, sharks (Carcharodon carcharias) in the western North Atlantic Ocean. Sci. Rep. R. (2011). Residency of the scalloped hammerhead shark (Sphyrna lewini) 8:10794. doi: 10.1038/s41598-018-29180-5 at Malpelo Island and evidence of migration to other islands in the Eastern Curtis, T. H., Zeeman, S. I., Summers, E. L., Cadrin, S. X., and Skomal, G. B. (2014). Tropical Pacific. Environ. Biol. Fishes 91, 165–176. doi: 10.1007/s10641-011- Eyes in the sky: linking satellite oceanography and biotelemetry to explore 9769-3 habitat selection by basking sharks. Anim. Biotelemetry 2:12. doi: 10.1186/2050- Block, B. A., Jonsen, I. D., Jorgensen, S. J., Winship, A. J., Shaffer, S. A., Bograd, S. J., 3385-2-12 et al. (2011). Tracking apex marine predator movements in a dynamic ocean. Dawe, E., Hendrickson, L., Colbourne, E., Drinkwater, K., and Showell, M. Nature 475, 86. doi: 10.1038/nature10082 (2007). Ocean climate effects on the relative abundance of short-finned Bornatowski, H., Braga, R., Abilhoa, V., and Corrêa, M. (2014). Feeding ecology (Illex illecebrosus) and long-finned (Loligo pealeii) squid in the northwest and trophic comparisons of six shark species in a coastal ecosystem off southern Atlantic Ocean. Fish. Oceanogr. 16, 303–316. doi: 10.1111/j.1365-2419.2007.00 Brazil. J. Fish Biol. 85, 246–263. doi: 10.1111/jfb.12417 431.x Braccini, M., Van Rijn, J., and Frick, L. (2012). High post-capture survival for Deacy, B. M., Moncrief-Cox, H. E., and Carlson, J. K. (2020). First verified record sharks, rays and chimaeras discarded in the main shark fishery of Australia?. of the smooth hammerhead (Sphyrna zygaena) in coastal waters of the northern PLoS One 7:e32547. doi: 10.1371/journal.pone.0032547 gulf of mexico with a review of their occurrence in the Western North Atlantic Breed, G. A., Jonsen, I. D., Myers, R. A., Bowen, W. D., and Leonard, M. L. (2009). Ocean. South. Nat. 19, 1–7. doi: 10.1656/058.019.0105 Sex-specific, seasonal foraging tactics of adult grey seals (Halichoerus grypus) Diemer, K., Mann, B., and Hussey, N. (2011). Distribution and movement of revealed by state–space analysis. Ecology 90, 3209–3221. doi: 10.1890/07- scalloped hammerhead Sphryna lewini and smooth hammerhead Sphyrna 1483.1 zygaena sharks along the east coast of southern Africa. Afr. J. Mar. Sci. 33, Calenge, C. (2006). The package adehabitat for the R software: a tool for the 229–238. doi: 10.2989/1814232X.2011.600291 analysis of space and habitat use by animals. Ecol. Modell. 197, 516–519. doi: Doyle, T. K., Bennison, A., Jessopp, M., Haberlin, D., and Harman, L. A. (2015). 10.1016/j.ecolmodel.2006.03.017 A dawn peak in the occurrence of ‘knifing behaviour’ in blue sharks. J. Anim. Cardeñosa, D., Fields, A. T., Babcock, E. A., Zhang, H., Feldheim, K., Shea, Biotelemetry 3:46. doi: 10.1186/s40317-015-0084-1 S. K., et al. (2018). CITES-listed sharks remain among the top species Dulvy, N. K., Fowler, S. L., Musick, J. A., Cavanagh, R. D., Kyne, P. M., Harrison, in the contemporary fin trade. Conserv. Lett. 11:e12457. doi: 10.1111/conl. L. R., et al. (2014). Extinction risk and conservation of the world’s sharks and 12457 rays. eLife 3:e00590. doi: 10.7554/eLife.00590 Casper, B., Domingo, A., Gaibor, N., Heupel, M., Kotas, E., Lamonaca, A., et al. Eddy, C., Brill, R., and Bernal, D. (2016). Rates of at-vessel mortality (2009). Sphyrna zygaena. The IUCN Red List of Threatened Species. Available a: and post-release survival of pelagic sharks captured with tuna purse http://www.iucnredlist.org (Accessed August 16, 2018). seines around drifting fish aggregating devices (FADs) in the equatorial Chamberlain, S., Tupper, B., and Mendelssohn, R. (2019). rerddap: General Purpose eastern Pacific Ocean. Fish. Res. 174, 109–117. doi: 10.1016/j.fishres.2015. Client for ’ERDDAP’ Servers, R package version 0.4.2.9130. 09.008 Clarke, S., Coelho, R., Francis, M., Kai, M., Kohin, S., Liu, K., et al. Ferretti, F., Worm, B., Britten, G. L., Heithaus, M. R., and Lotze, H. K. (2010). (2015). Report of the Pacific Shark Life History Expert Panel Workshop: Patterns and ecosystem consequences of shark declines in the ocean. Ecol. Lett. Final Report. Kolonia: Western and Central Pacific Fisheries Commission 13, 1055–1071. doi: 10.1111/j.1461-0248.2010.01489.x Scientific Committee. Francis, M. (2016). Distribution, habitat and movement of juvenile smooth Clarke, S. C., Magnussen, J. E., Abercrombie, D. L., McAllister, M. K., and Shivji, hammerhead sharks (Sphyrna zygaena) in northern New Zealand. M. S. (2006a). Identification of shark species composition and proportion N. Zeal. J. Mar. Freshw. Res. 50, 506–525. doi: 10.1080/00288330.2016.117 in the Hong Kong shark fin market based on molecular genetics and 1244 trade records. Conserv. Biol. 20, 201–211. doi: 10.1111/j.1523-1739.2005.00 Freitas, C., Lydersen, C., Fedak, M. A., and Kovacs, K. M. (2008). A simple new 247.x algorithm to filter marine mammal Argos locations. Mar. Mamm. Sci. 24, Clarke, S. C., McAllister, M. K., Milner-Gulland, E. J., Kirkwood, G., Michielsens, 315–325. doi: 10.1111/j.1748-7692.2007.00180.x C. G., Agnew, D. J., et al. (2006b). Global estimates of shark catches using trade Gallagher, A. J., and Klimley, A. P. (2018). The biology and conservation status records from commercial markets. Ecol. Lett. 9, 1115–1126. doi: 10.1111/j. of the large hammerhead shark complex: the great, scalloped, and smooth 1461-0248.2006.00968.x hammerheads. Rev. Fish Biol. Fish. 28, 777–794. doi: 10.1007/s11160-018- CLS (2016). Argos Users Manual. Available online at: http://www.argos-system. 9530-5 org/wp-content/uploads/2016/08/r363_9_argos_users_manual-v1.6.6.pdf Grubbs, R. D. (2010). “Ontogenetic shifts in movements and habitat use,” in (accessed September 15, 2018). Sharks and Their Relatives II, eds M. Heithaus, J. C. Carrier, and J. A. Coelho, R., Fernandez-Carvalho, J., Lino, P. G., and Santos, M. N. (2012). An Musick (Boca Raton, FL: CRC press), 335–366. doi: 10.1201/9781420080 overview of the hooking mortality of elasmobranchs caught in a swordfish 483-c7

Frontiers in Marine Science| www.frontiersin.org 10 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 11

Logan et al. Smooth Hammerhead Shark Seasonal Movement

Grubbs, R. D., Musick, J. A., Conrath, C. L., and Romine, J. G. (2007). “Long- N. Kohler, and H. Pratt (Bethesda, MD: American Fisheries Society), term movements, migration, and temporal delineation of a summer nursery 45–62. for juvenile sandbar sharks in the Chesapeake Bay region,” in Proceedings of Mendelssohn, R. (2017). Package ’xtractomatic: Accessing Environmental the American Fisheries Society Symposium, (Bethesda, MD: American Fisheries Data from ERD’s ERDDAP Server’. Vienna: R Foundation for Society). Statistical Computing. Guttridge, T. L., Van Zinnicq Bergmann, M. P., Bolte, C., Howey, L. A., Finger, Miller, M. H. (2016). Endangered Species Act Status Review Report: Smooth J. S., Kessel, S. T., et al. (2017). Philopatry and regional connectivity of the great Hammerhead Shark (Sphyrna zygaena). Silver Spring, MD: National Marine hammerhead shark, Sphyrna mokarran in the US and Bahamas. Front. Mar. Sci. Fisheries Service, Office of Protected Resources. 4:3. doi: 10.3389/fmars.2017.00003 Morgan, A., and Carlson, J. K. (2010). Capture time, size and hooking mortality Heyman, W. D., Graham, R. T., Kjerfve, B., and Johannes, R. E. (2001). Whale of bottom longline-caught sharks. Fish. Res. 101, 32–37. doi: 10.1016/j.fishres. sharks Rhincodon typus aggregate to feed on fish spawn in Belize. Mar. Ecol. 2009.09.004 Prog. Ser. 215, 275–282. doi: 10.3354/meps215275 Mourier, J., Maynard, J., Parravicini, V., Ballesta, L., Clua, E., Domeier, M. L., Hijmans, R. J., van Etten, J., Cheng, J., Mattiuzzi, M., Sumner, M., et al. (2016). Extreme inverted trophic pyramid of reef sharks supported Greenberg, J. A., et al. (2017). Package ‘Raster’. Vienna: R Foundation for by spawning groupers. Curr. Biol. 26, 2011–2016. doi: 10.1016/j.cub.2016. Statistical Computing. 05.058 Hoffmayer, E. R., Franks, J. S., Driggers, W. B., and Howey, P. W. (2013). Musick, J., and Colvocoresses, J. (1986). “Seasonal recruitment of subtropical Diel vertical movements of a scalloped hammerhead, Sphyrna lewini, in the sharks in Chesapeake Bight, USA,” in Workshop on Recruitment in Tropical northern Gulf of Mexico. Bull. Mar. Sci. 89, 551–557. doi: 10.5343/bms.2012. Coastal Demersal Communities. Intergovenmental Oceanographic Commission 1048 Workshop Report 44, eds A. Yanez, Y. Arancibia, and D. Pauley (Campeche: Hothorn, T., Bretz, F., and Westfall, P. (2008). multcomp: simultaneous inference FAO/UNESCO), 301–311. in general parametric models. R package version 1.4-6. Biometr. J. 50, 346–363. Nalesso, E., Hearn, A., Sosa-Nishizaki, O., Steiner, T., Antoniou, A., Reid, A., et al. doi: 10.1002/bimj.200810425 (2019). Movements of scalloped hammerhead sharks (Sphyrna lewini) at Cocos Jereb, P., and Roper, C. (2010). Cephalopods of the world. An annotated Island, Costa Rica and between oceanic islands in the Eastern Tropical Pacific. and illustrated catalogue of species known to date. Volume 2. Myopsid PLoS One 14:e0213741. doi: 10.1371/journal.pone.0213741 and Oegopsid Squids. Rome: Food and Agriculture Organization of the NMFS (2009). Final Amendment 1 to the 2006 Consolidated Atlantic Highly United Nations. Migratory Species Fishery Management Plan, Essential Fish Habitat. Silver Jiao, Y., Cortés, E., Andrews, K., and Guo, F. (2011). Poor-data and data-poor Spring, MD: NMFS. species stock assessment using a Bayesian hierarchical approach. Ecol. Applic. Priede, I. G., and Miller, P. I. (2009). A basking shark (Cetorhinus maximus) tracked 21, 2691–2708. doi: 10.1890/10-0526.1 by satellite together with simultaneous remote sensing II: new analysis reveals Jonsen, I. (2016). Joint estimation over multiple individuals improves behavioural orientation to a thermal front. Fish. Res. 95, 370–372. doi: 10.1016/j.fishres. state inference from animal movement data. Sci. Rep. 6:20625. doi: 10.1038/ 2008.09.038 srep20625 Queiroz, N., Humphries, N. E., Mucientes, G., Hammerschlag, N., Lima, F. P., Jonsen, I., Bestley, S., Wotherspoon, S., Sumner, M., and Flemming, J. M. (2015). Scales, K. L., et al. (2016). Ocean-wide tracking of pelagic sharks reveals extent bsam: Bayesian State-Space Models For Animal Movement. R package. R of overlap with longline fishing hotspots. Proc. Natl. Acad. Sci. U.S.A. 113, Foundation for Statistical Computing version 0.43. 1. 1582–1587. doi: 10.1073/pnas.1510090113 Jonsen, I. D., Flemming, J. M., and Myers, R. A. (2005). Robust state–space R Core Team (2014). R: A Language and Environment for Statistical Computing. modeling of animal movement data. Ecology 86, 2874–2880. doi: 10.1890/04- Vienna: R Foundation for Statistical Computing. 1852 Rogers, P. J., Huveneers, C., Page, B., Hamer, D. J., Goldsworthy, S. D., Mitchell, Kajiura, S. M., and Tellman, S. L. (2016). Quantification of massive seasonal J. G., et al. (2012). A quantitative comparison of the diets of sympatric pelagic aggregations of blacktip sharks (Carcharhinus limbatus) in Southeast Florida. sharks in gulf and shelf ecosystems off southern Australia. ICES J. Mar. Sci. 69, PLoS One 11:e0150911. doi: 10.1371/journal.pone.0150911 1382–1393. doi: 10.1093/icesjms/fss100 Ketchum, J. T., Hearn, A., Klimley, A. P., Peñaherrera, C., Espinoza, E., Bessudo, Santos, C. C., and Coelho, R. (2018). Migrations and habitat use of the smooth S., et al. (2014). Inter-island movements of scalloped hammerhead sharks hammerhead shark (Sphyrna zygaena) in the Atlantic Ocean. PLoS One (Sphyrna lewini) and seasonal connectivity in a marine protected area of the 13:e0198664. doi: 10.1371/journal.pone.0198664 eastern tropical Pacific. Mar. Biol. 161, 939–951. doi: 10.1007/s00227-014- Santos, C. C., and Coelho, R. (2019). Distribution patterns and indicators of the 2393-y smooth hammerhead shark (Sphyrna zygaena) in the Atlantic Ocean. Fish. Res. Kinney, M. J., and Simpfendorfer, C. A. (2009). Reassessing the value of nursery 212, 107–113. doi: 10.1016/j.fishres.2018.12.015 areas to shark conservation and management. Conserv. Lett. 2, 53–60. doi: Schlaff, A. M., Heupel, M. R., and Simpfendorfer, C. A. (2014). Influence of 10.1111/j.1755-263x.2008.00046.x environmental factors on shark and ray movement, behaviour and habitat Klimley, A. P., Anderson, S. D., Pyle, P., and Henderson, R. (1992). use: a review. Rev. Fish Biol. 24, 1089–1103. doi: 10.1007/s11160-014- Spatiotemporal patterns of white shark (Carcharodon carcharias) predation at 9364-8 the South Farallon Islands, California. Copeia 1992, 680–690. doi: 10.2307/ Smale, M. (1991). Occurrence and feeding of three shark species, Carcharhinus 1446143 brachyurus, C. obscurus and Sphyrna zygaena, on the Eastern Cape coast of Klimley, A. P., and Nelson, D. R. (1984). Diel movement patterns of the scalloped South Africa. South Afr. J. Mar. Sci. 11, 31–42. doi: 10.2989/02577619178428 hammerhead shark (Sphyrna lewini) in relation to El Bajo Espiritu Santo: a 7808 refuging central-position social system. Behav. Ecol. Sociobiol. 15, 45–54. doi: Spaet, J. L., Lam, C. H., Braun, C. D., and Berumen, M. L. (2017). Extensive 10.1007/bf00310214 use of mesopelagic waters by a Scalloped hammerhead shark (Sphyrna Kohler, N. E., and Turner, P. A. (2019). Distributions and movements of atlantic lewini) in the Red Sea. Anim. Biotelemetry 5:20. doi: 10.1186/s40317-017- shark species: a 52-year retrospective atlas of mark and recapture data. Mar. 0135-x Fish. Rev. 81, 1–93. doi: 10.7755/mfr.81.2.1 Teter, S. M., Wetherbee, B. M., Fox, D. A., Lam, C. H., Kiefer, D. A., and Shivji, M. Lea, J. S., Wetherbee, B. M., Sousa, L. L., Aming, C., Burnie, N., Humphries, N. E., (2015). Migratory patterns and habitat use of the sand tiger shark (Carcharias et al. (2018). Ontogenetic partial migration is associated with environmental taurus) in the western North Atlantic. Mar. Freshw. Res. 66, 158–169. doi: drivers and influences fisheries interactions in a marine predator. ICES J. Mar. 10.1071/mf14129 Sci. 75, 1383–1392. doi: 10.1093/icesjms/fsx238 Vaudo, J. J., Byrne, M. E., Wetherbee, B. M., Harvey, G. M., and Shivji, M. S. McCandless, C., Pratt, H., Kohler, N., Merson, R., and Recksiek, C. (2007). (2017). Long-term satellite tracking reveals region-specific movements of “Distribution, localized abundance, movements, and migrations of juvenile a large pelagic predator, the shortfin mako shark, in the western North sandbar sharks tagged in Delaware Bay,” in Shark Nursery Grounds of the Gulf Atlantic Ocean. J. Appl. Ecol. 54, 1765–1775. doi: 10.1111/1365-2664. of Mexico and the East Coast Waters of the United States, eds C. McCandless, 12852

Frontiers in Marine Science| www.frontiersin.org 11 September 2020| Volume 7| Article 566364 fmars-07-566364 August 30, 2020 Time: 10:2 # 12

Logan et al. Smooth Hammerhead Shark Seasonal Movement

Ware, D. M., and Thomson, R. E. (2005). Bottom-up ecosystem trophic dynamics Conflict of Interest: MS was employed by the company Fish Finder Adventures. determine fish production in the Northeast Pacific. Science 308, 1280–1284. doi: 10.1126/science.1109049 The remaining authors declare that the research was conducted in the absence of Wells, R., TinHan, T. C., Dance, M. A., Drymon, J. M., Falterman, B., Ajemian, any commercial or financial relationships that could be construed as a potential M. J., et al. (2018). Movement, behavior, and habitat use of a marine apex conflict of interest. predator, the scalloped hammerhead. Front. Mar. Sci. 5:321. doi: 10.3389/fmars. 2018.00321 Copyright © 2020 Logan, Vaudo, Sousa, Sampson, Wetherbee and Shivji. This is an Weng, K. C., Foley, D. G., Ganong, J. E., Perle, C., Shillinger, G. L., and Block, open-access article distributed under the terms of the Creative Commons Attribution B. A. (2008). Migration of an upper trophic level predator, the salmon shark License (CC BY). The use, distribution or reproduction in other forums is permitted, Lamna ditropis, between distant ecoregions. Mar. Ecol. Prog. Ser. 372, 253–264. provided the original author(s) and the copyright owner(s) are credited and that the doi: 10.3354/meps07706 original publication in this journal is cited, in accordance with accepted academic Zuur, A., Ieno, E., Walker, N., Saveliev, A., and Smith, G. (2009). Mixed Effects practice. No use, distribution or reproduction is permitted which does not comply Models and Extensions in Ecology with R. New York, NY: Springer. with these terms.

Frontiers in Marine Science| www.frontiersin.org 12 September 2020| Volume 7| Article 566364